Python Multinested Dictonary To Pandas Dataframe Boolean Table-CodePudding

I am looking for any advice on how to cleanly convert a python multi level nested dictionary (from JSON) into a data frame boolean table.

Rules:

Only True is recorded. If empty it is False.
The list may be of N Length
The groups may be of N length
The bools may be of N types

Example Input:

{1:{group_a:{bool_a:True,
             bool_b:True,
             bool_n:True},
    group_n:{bool_b:True,
             bool_n:True}
   },
2:{group_a:{bool_a:True,
             bool_b:True,
             bool_n:True},
    group_n:{bool_b:True,
             bool_n:True}
   },
n:{group_a:{bool_a:True,
            bool_c:True},
    group_n:{bool_b:True}
   },
}

Desired Output:

   Ga_Ba, Ga_Bb, Ga_Bc, Ga_Bn, Gn_Ba, Gn_Bb, ... Gn_Bn....
1   True   True  False  True   False   True       True
2   True   True  False  True   False   True       True
n   True   False True   False  False   False      False
...

Ideas? Bonus points for speed and conciseness. I have a solution but I am looking for something more elegant than the for loop mess I have now. Alternative data structures may also be welcome.

CodePudding user response：

Goofy method #1

s = pd.DataFrame.from_dict(data, orient='index').stack()

pd.json_normalize(s).set_index(s.index) \
  .stack().unstack([1, 2], fill_value=False) \
  .sort_index(axis=1)

  group_a                      group_n       
   bool_a bool_b bool_c bool_n  bool_b bool_n
1    True   True  False   True    True   True
2    True   True  False   True    True   True
3    True  False   True  False    True  False

CodePudding user response：

You could use a dictionary comprehension and concat:

df = (pd.concat({k: pd.DataFrame(v).stack()
                 for k,v in d.items()}, axis=1)
        .T.fillna(False)
      )